Home: Program Supplement

INTERVIEW WITH DR WILLIAM SEE

DR NEIL LOVE: Later on in the program, Dr Dawson will present two cases from her practice, but one key paper she identified was from the massive Early Prostate Cancer trial comparing bicalutamide to placebo. I met with Dr William See, a principle investigator on this groundbreaking research effort to learn more of the early results that have been observed, and he began by providing an overview of how this research project was organized.

DR WILLIAM SEE: There were three trials. There was the North American trial, which was Canada and the United States. There was the Capri trial, which really was Europe and what I'll say is the rest of the world. It included South America, Central America, Europe and Australia. And then there was the SPCG trial, which was a Swedish trial.

So the program was designed for a pooled analysis to encompass all of the patients enrolled in those three trials. To accomplish that, we used identical end points for the trial, and so the two primary end points for the trial were objective disease progression, not PSA doubling, but objective disease progression and survival. We wanted patients in all three sites to have non-metastatic prostate cancer, and so you could have locally advanced disease, but you couldn't have bony metastases.

Now, there were some differences between the three trial sites, and these differences were allowed because of some cultural differences in how the disease was being managed, at that time, in each of those countries. And I think Scandinavia illustrates the issues most clearly. In Scandinavia, watchful waiting was probably the preferred approach when the trial was initiated. So not to allow inclusion of those patients would really have disadvantaged accrual at those sites. In addition, I think that the watchful waiting patients will probably provide some very meaningful data about the concept of early hormonal therapy in patients who are not receiving primary therapy of curative intent. I think that will prove to be beneficial. So, we allowed that in Scandinavia. That wasn't being done in this country regularly, and so patients who were being watchfully waited could not be included if they were in North America.

The inclusion of node-positive patients was another difference between the different clinical trial sites. In North America, you could not have node-positive disease, whereas in Europe and Scandinavia, you could. And I've already alluded to the fact, sort of post hoc, we allowed some differences in the duration of therapy. In North America, we limited the duration of treatment to two years. In Europe, it was allowed to be five years. And in Scandinavia, it could be indefinite.

I think it's important to preface that discussion by what have proven to be some differences in risk for disease progression, or PSA doubling, as a function of the different trials. As we look at PSA as a surrogate for disease volume across the three trials, we see some very distinct differences. In the North American trial, the average PSA at randomization was seven. In contrast to that, the average PSA in Scandinavia was more than twice that. And so, as we look back, these really are very different populations for lots of reasons, but predominantly by extent of disease. And as we now get data as it relates to risk of progression, we're seeing a risk of progression that correlates with disease extent. In other words, the event rate is being driven by the priority of risk. The best risk population was in North America. We don't see a difference, at this point, in time in terms of the risk of objective progression.

DR LOVE: Now, at this point, which is just within about two years of follow-up?

DR SEE: Correct. I think it's five to seven. It's very low. Again, it's the best group. And it reflects what's happened in this country. We've seen a huge stage shift. We're detecting it earlier and earlier. These, in fact, were great risk patients. They were going to do well for a long period of time no matter what you did. And it's that event rate, the relatively low event rate, that I think underlies the reason we haven't been able to interpret the data at this point in time.

Conversely, in the group that has the highest predicted event rate, based upon disease extent, that's where we've really seen the curves separate. So in Scandinavia and, to some degree, in Europe, there is a highly significant difference in risk of progression between the placebo and the bicalutamide group in favor of the adjuvant anti-androgen.

One of the interesting things about the data may be the insight that PSA doubling provides us. Although the FDA doesn't necessarily agree, I do believe that PSA is a surrogate for outcome, and it will ultimately predict for the likelihood of either disease progression or risk for death from disease. In North America, although we don't see any difference at this point in time in terms of objective progression, what we do see is a significant difference in the risk of PSA doubling. And so if you think that PSA is a harbinger of things to come, then, as time goes on and we do get more events, then we would anticipate that the curves for progression in North American will begin to separate.

DR LOVE: Another issue that might have to be considered is in North America, the duration was a little bit shorter, being two years compared to five or indefinite. So, theoretically, that could be an explanation. But you're saying that most likely, looking at the numbers, it's just a lack of events at this point.

DR SEE: Too few events at this point is my interpretation.

DR LOVE: Is that the same explanation you have in terms of survival? What are you seeing right now in terms of survival with the three trials?

DR SEE: Somewhere around five percent of patients in the overall trial have died. If we're looking at time to median survival, we've got a long way to go. It is simply too early. One of the things I spend a lot of my time doing is, I speak to what I think is a very important trial, is clarifying for the audience that this trial has not stopped. This trial is, in fact, ongoing, and that we continue to follow these patients for both objective progression and, more importantly, for survival. And so although I don't have an answer for you today, at some point in the future, we will have an answer about that critical end point.

DR LOVE: What do you see when you look at the effects of therapy based on the local treatment?

DR SEE: For primary treatment modality, what we have seen is that irrespective of whether you were treated with radical prostatectomy, with radiotherapy or watchful waiting, there was a persistent significant benefit in reducing the risk of disease progression associated with bicalutamide. In other words, it didn't matter what you were treated with, if you received bicalutamide, you saw a benefit.

DR LOVE: Now, is this trial going to evaluate what the benefit would be of waiting until PSA is increased compared to not?

DR SEE: Well, I think there's a tacit assumption that patients in the placebo arm, as they progress, will have standard-care therapy. So those patients will ultimately get hormonal therapy, although that is not mandated in the trial. What patients are allowed to do at the time of progression is left to the discretion of the treating physician.

In this trial, PSA was not an end point. And so, if your PSA doubled, as an example, that was not a protocol-driven cause to come off study and go onto second-line therapy. But patients, obviously, could elect to withdraw from the trial and be treated with some second-line treatment modality.

DR LOVE: Could you summarize what you would say to a patient who asks what is the potential benefit of this intervention?

DR SEE: What I would suggest for patients who are at high risk for disease progression is to tell them about the data. And there probably is a line beyond which you're going to say, well, this is a reasonable treatment strategy, versus it may not be. Now, should you leave that up to the patient? Should you, in fact, just give them the data and let them make the decision? I think to be honest, you probably should. I think the right thing to do would be to probably have the discussion and let the patient enter the data into their own personalized risk-benefit ratio balance and say, well, that's not for me, or this is for me.

But what I would tell patients related to this study is that we've had a primary therapy in America. That's the majority of the patients I'm seeing. So, you've had either radiation or radical prostatectomy. We know that those treatment modalities are not perfect. We know that, irrespective of our best intent, you have a risk of treatment failure, which in your case might be X. We have some data that now suggests we can reduce at least the risk of objective progression in PSA doubling in patients with non-metastatic prostate cancer through the use of this adjuvant anti-androgen bicalutamide.

What are the down sides of that therapy going to be? Well, the majority of patients are going to have gynecomastia and breast pain. We haven't talked yet about the severity of that, the reversibility of that or strategies to prevent that, but that would need to be part of the discussion with patients. Patients are going to need to know that probably 70-80 percent of patients may have one or both of those symptoms.

The final thing we need to be telling patients is that at this juncture, we don't know whether this reduction in the risk of PSA doubling or objective progression translates into a survival advantage. Because for prostate cancer, what we have struggled with for years has been the argument between early versus delayed hormonal therapy. We know we've got this treatment strategy in our pocket. Charlie Huggins showed us in the 1940s that if you withdraw androgens from patients with advanced prostate cancer, you can have a remarkable benefit.

What we have gone back and forth about for now over 60 years is: When do we do that? Do we do it very early? Do we wait until the patient becomes symptomatic? I think, historically, certainly, myself included, 10 years ago, I was in the mindset of delayed hormonal therapy. The data at that point in time suggested that it didn't matter whether you started hormonal therapy when you were asymptomatic today or whether you waited until you started to have symptoms. The time at which you were going to succumb to the disease was unchanged. And the only difference between those two treatment strategies was the duration of toxicity that you incurred as a consequence of the active agent.

Many patients would say, I feel fine now. Why should I take something, if I'm not going to live any longer? I think what we've seen over the last decade really has been a growing body of evidence to suggest that earlier intervention may, in fact, be associated with a survival advantage.

We have Ed Messing's study published in the New England Journal in December of 1999. It looked at patients with node-positive prostate cancer, who had undergone a radical prostatectomy. They were randomized to receive immediate androgen ablation or androgen ablation at the time of disease progression. What we saw was a highly statistically significant difference in survival.

There have been a couple of other trials that have looked at the use of adjuvant hormonal therapy associated with either radiation therapy or with radical prostatectomy. The radical prostatectomy data is relatively immature, but I think the radiotherapy data is very intriguing. You look at the Bolla trial, where they took patients with relatively clinically advanced prostate cancer, undergoing primary radiotherapy. Randomized them to three years of adjuvant hormonal therapy versus nothing. Not only did the adjuvant hormonal therapy prevent or reduce the risk of progression, but that actually translated into a survival advantage in those patients.

So, those issues are prompting many of us to reassess our historic stance on the timing of implementation of hormonal therapy. I personally have shifted towards earlier rather than delayed, recognizing that there is still a big gap that's missing as to when, in fact, to do that. I think many urologists justifiably are going to be asking themselves: In the absence of survival data, should I be telling my patients about this? It gets back to where we were with breast 15 years ago. We were in the identical setting. We had data about adjuvant anti-estrogens in breast that said progression was benefited, but we didn't have survival. What did we do?

Well, there's obviously some uncertainty, but I think that we are compelled to at least tell patients about this data and let them make a decision about whether or not they want to engage in this treatment strategy, with a clear understanding of the risks and benefits in the absence of survival data.

DR LOVE: It's interesting, too, when you look at the example of breast, that while it's true that researchers are always looking for a survival difference, what we've seen was that patients themselves valued not having disease progression. You would wonder whether you would see the same thing in men with prostate cancer and whether or not even men with prostate cancer would value not having PSA progression.

DR SEE: Well, one of the events that I fear most in my clinical practice is the discussion I have with patients when their PSA comes back after some definitive therapy. You know, their face goes ashen. They're devastated. They know what that implies. And although I try and salvage them, I go, your average life expectancy from this point forward is in excess of 10 years, they're selling the farm. They're ready to cash it in at that point in time.

And so, is there, in fact, a value from the patient's perspective in delaying that from occurring? I don't know. One of the great mistakes I think we made in the EPC trial was not prospectively collecting quality-of-life data. But I do think that we need additional insight into the patients' perception of end points like progression, like absence of biochemical failure, even if survival is never impacted. Obviously, those have got to be considered in the context of the down sides associated with adjuvant hormonal therapy.

DR LOVE: Also, you could envision a situation, perhaps, where the survival was not affected, but yet the period of time being disabled or ill might be decreased.

DR SEE: Let's take an example of the MRC trial. The MRC trial looked at this issue of early versus delayed hormonal therapy in men with more advanced prostate cancer not receiving primary therapy of curative intent. And although it's been analyzed lots of ways and it's been highly criticized, I think the one thing that's clear from that trial is that there was a reduction in the risk of catastrophic disease progression associated with earlier implementation. And so fewer pathologic fractures, fewer cord-compromise events associated with earlier therapies. And that's something else that we need to factor in.

The only other comment I would make is in relationship to the robustness of the benefit, that try as we might, we have yet to identify a group that was disadvantaged by this treatment strategy, and so we see benefit virtually irrespective of primary treatment modality, node positivity, local disease extent, Gleason score. PSA should be above four, but beyond that, there's benefit. So, this is very robust data, and I personally am very excited by it and very proud to have been a participant in this trial. But I do think we need to continue to follow these patients to see if this will ultimately translate into survival.

DR LOVE: Can you talk a little bit about what the side effects and tolerability profile were in the study?

DR SEE: Tolerability differed a little bit by site, first of all. What I mean by that is withdrawals due to adverse events differed among the three different trial sites. So, they were relatively low in the Scandinavian trial, somewhere in the range of three percent of patients on the active treatment withdrew from drug. That contrasts with this country, where somewhere in the range of 17 percent of patients withdrew from drug due to adverse events. So, there was a difference in cultural perception, or tolerance for those adverse events, as a function of trial center. What were the principle adverse events that prompted patients to withdraw irrespective of site? Well, they related to gynecomastia and breast tenderness.

There's an interesting pharmacology about this drug that causes a relatively high incidence of gynecomastia, upwards of 70 percent in this trial. If you recall that the anti-androgens do not, in fact, decrease serum levels of testosterone. In fact, because they are blocking the pituitary hypothalamic perception of testosterone levels, there's actually an increased production of LH and an increased testosterone synthesis by the testicles. What the anti-androgens do, is block the androgen receptor at the level of the cell to preclude circulating levels of testosterone from interacting with the androgen receptor and stimulating cell growth and other cell biologic features.

So, you see in patients receiving bicalutamide monotherapy an increase in serum testosterone. What happens to that serum testosterone? Well, the liver and fat can convert it to estrogenic compounds, and so estrogenic-sensitive tissues, such as the breast, are stimulated by that increase in circulating levels of estrogens, resulting in the gynecomastia and breast tenderness that we see. And so the majority of patients on Casodex monotherapy in this trial experienced one or both of those symptoms.

Those were reversible to some degree. Breast pain in particular was reversible, and so following cessation of therapy, there was a very nice decay curve in patients with persistent breast pain. Gynecomastia, however, was somewhat less reversible, and so, although some patients did see a reduction in breast volume - I want to say of those who stopped drug, perhaps 50-60 percent still had persistence of gynecomastia a year or two out from cessation of therapy. So patients are going to need to be aware of that.

Are there ways to prevent that, I think, is one of the issues we've been confronted with. Should we be doing those? I don't know. DES - had a very similar side effect. And patients were irradiated in the '60s and '70s prior to receiving DES. There was, in fact, remarkable objective literature about when to give radiotherapy or what doses to give from that era. What we did know is that radiotherapy was far more effective given prophylactically then therapeutically. And so the response of the developed gynecomastia to breast irradiation is relatively poor.

I do know of a clinical trial that's currently underway. Chris Tyrrell in London is looking at a short - I believe it's a single-fraction radiation therapy to prevent the development of gynecomastia. His treatment regimen was about 50 percent successful in preventing the development. And that was specifically for gynecomastia. I think that radiotherapy is even more effective for the breast pain, but it's a little unclear as to just how successful it will be for the breast enlargement issue.

DR SEE: There's a lot of variability in my patients, between their acceptance of the gynecomastia. I think that for young, active men, they found it troubling. Conversely, the older the man, sort of the larger the body habitus perhaps. The greater the degree of apriorie gynecomastia, the less problematic it was. So, some patients will perceive this as something to be avoided at all costs. Others probably will be relatively indifferent. The other thing that we need to reflect on from the data in the EPC trial is why are there these differences between the different clinical sites in terms of tolerability of adverse events? Why were Scandinavians much more tolerant than North Americans, related to this issue of gynecomastia and breast pain?

To some degree, it could be cultural. But, more likely, I think it's related to a perceived benefit. The majority of patients in Scandinavia were viewing their hormonal therapy as their primary treatment modality. So, the majority of patients in that site were on watchful waiting, and so the hormonal therapy was their treatment modality. That was what they were going to see benefit from.

Conversely, in North America, these were patients who'd already theoretically been cured of their prostate cancer. Although they were at risk for disease progression, the benefit associated with this at that time, experimental treatment strategy, was far less clear. Consequently, I suspect their willingness to accept toxicity was lower.

DR LOVE: Now, you talk about breast pain, but is this breast pain or breast tenderness?

DR SEE: Well, it's semantics, I guess, at some level. It's breast tenderness. Most patients don't complain of constant pain. There's sensitivity to areolae and the sub-areolae tissue. So, if their shirt rubs against their breasts, they notice it. So, it's not constant pain. It's more of an induced phenomenon.

DR LOVE: Again, how much of a problem did you perceive that to be?

DR SEE: The patients were, in fact, more tolerant of the breast pain than the breast enlargement, the breast discomfort or breast tenderness, depending upon what you want to call it. It was easier to tell them that that was going to go away following cessation of therapy, as the data, in fact, shows. Conversely, the reversibility of the gynecomastia not only was anticipated to be less, but also, in fact, proved to be so.

DR LOVE: What other symptoms or side effects are seen?

DR SEE: We did see some hot flushes in this trial. That's predictable. And the sexual function and decreased libido issue was specifically addressed by a questionnaire in the Scandinavian trial. There was a reduction in feeling sexual or interest in sex associated with bicalutamide, but compared to the placebo group the difference was relatively small. My interpretation of the data is that it's a far smaller reduction in libido than would have been seen had you had a conventional agent such as an LHRH agonist.

DR LOVE: Let's just say, for example, it had been a three-arm trial, and the third arm was an LHRH agonist. What would you have expected granted that you don't know the answer? But by looking at the data indirectly, what would you have expected in terms of anti-tumor effect?

DR SEE: My sense is that it would be the same, and that is driven by additional trials that have looked at the 150-milligram dose of bicalutamide. There have been a number of those. There was a trial that looked at 150 milligrams versus an LHRH agonist for patients with metastatic prostate cancer. There was actually a difference in survival in that particular study. The difference, the average duration of survival, I want to say, was 42 days between the groups.

But there have been other trials in earlier-stage disease that suggests that, in fact, there's no difference. So I think it's going to be a function of tumor burden and, as you shift earlier and earlier in the natural history of the disease, that anti-androgens may well prove to be equivalent to some of the other agents, with less toxicity. And, in fact, it's that balance of benefit versus risk that drives our selection of treatment modalities when given in the adjuvant context.

DR LOVE: What about effects on bone?

DR SEE: That's a very interesting question, as well. I go around and I give a lecture about what I call the dark side of hormonal therapy because all of these agents have some risk associated with their use. The classic things for hormonal therapy have been loss of muscle mass, loss of bone mineral density, perhaps cognitive impairment, loss of sexual function. These are real quality-of-life things for aging men.

Does Casodex prevent that? Well, there is some very preliminary data to suggest that there's no difference in terms of changes in bone mineral density over time in patients receiving the 150-milligram dose of Casodex versus placebo.

Conversely, there is a plethora of data that show that LHRH agonist therapy is associated with a progressive loss of bone mineral density, which increases as a function of duration of therapy.

DR LOVE: What would be the mechanism for why Casodex wouldn't affect bone?

DR SEE: Well, we get into this issue of the circulating testosterone and is that, in fact, protective at the bone level? The whole science behind changes in bone mineral density is interesting. M hypothesis is that those circulating levels of testosterone, even though they may have less access to the androgen receptor still are functionally protective for the bone.

Home

Editor's Office